What Is the True Impact of Periprosthetic Joint Infection Diagnosis on Mortality?

Mortality is a quality indicator that may affect expenditures. Revisions for periprosthetic joint infection (PJI) are, on average, more expensive and exhibit higher morbidity than aseptic revisions, although reimbursement is similar. Therefore, we sought to determine (1) impact on mortality rates of revision total hip and/or knee arthroplasty performed for PJI diagnosis (septic) versus aseptic revisions, at any point in time, and (2) mortality predictors among PJI patients.

Retrospective chart review of 978 consecutive patients who underwent revision at a single institution (January 2015 to November 2020). All revisions were evaluated, and it was determined whether patients had a revision for PJI at any point in time or not. Two groups were compared: (1) patients with septic revision(s) (n = 350) and (2) patients who only underwent aseptic revision(s) (n = 628). Demographics and mortality status at latest follow-up (mean 3 years, range: 0 to 18 years, from first revision ever) were assessed. Mortality status was also separately assessed among patients who exclusively had hip revision(s), or solely knee revision(s), or both. Multivariate regression analysis (Cox) was used to determine whether PJI diagnosis was an independent mortality predictor. Among PJI patients, potential mortality predictors were evaluated.

Overall, 65 patients died (6.6%). The septic cohort had significantly more men and American Society of Anesthesiologists (ASA) class-IV patients. Mortality rates were 10.9% and 4.3% (P < .0001) for septic and aseptic revision groups, respectively. After controlling for sex, ASA, and number of revisions, PJI diagnosis was a significant mortality predictor (hazard ratio [HR]: 2.69, 95% confidence interval [CI]: 1.5 to 4.7, P = .001). Among PJI patients, age (HR: 1.05, 95% CI: 1.01 to 1.08, P = .009) and ASA (HR: 4.02, 95% CI: 1.67 to 9.67, P = .002) were independent predictors.

Having a revision due to PJI diagnosis was associated with 2.5 times increased mortality. Therefore, more accurate coding capturing the complexity and morbidity of revisions for PJI diagnosis is needed.